| opportunity |
location |
|
| 13.25.03.C0822 |
Wright-Patterson AFB, OH 454337817 |
The next generation of optical materials will utilize wavelength-scale geometric features to manipulate electromagnetic waves in ways beyond the capability of traditional bulk materials to achieve desired performance. Due to the near-infinite design space for such metamaterials, advanced optimization strategies coupled with computational electromagnetic simulation must be employed to accelerate the design process. Machine learning approaches have demonstrated the potential to overcome the limitations of ansatz and/or iterative approaches by inverting the design process, directly delivering metamaterial designs with the required properties. We will investigate the creation of machine learning models capable of designing metamaterials, the use of these models to design metamaterials with novel electromagnetic properties, and the fabrication of these designs. Candidates with backgrounds in optics, materials science & engineering, physics, or related fields with machine learning experience are highly encouraged to contact the adviser to discuss research opportunities.
[1] Inverse design of broadband highly reflective metasurfaces using neural networks, Eric S. Harper, Eleanor J. Coyle, Jonathan P. Vernon, Matthew S. Mills. Phys. Rev. B, 101, 195104, 4 May 2020. DOI: 10.1103/PhysRevB.101.195104
[2] Particle swarm optimization of polymer-embedded broadband metasurface reflectors, Jonathan R. Thompson, Heidi D. Nelson-Quillin, Eleanor J. Coyle, Jonathan P. Vernon, Eric S. Harper, Matthew S. Mills. Optics Express, 29, 26, 20 December 2021. DOI: 10.1364/OE.444112
[3] Diffractive deep neural network adjoint assist or (DNA)2: a fast and efficient nonlinear diffractive neural network implementation, Ighodalo U Idehenre, Eric S Harper, Matthew S Mills. Optics Express, 30, 5, 28 February 20222. DOI: 10.1364/OE.449415
Machine Learning; Metasurfaces; Metamaterials; Optics; Inverse Design; Material Science & Engineering; Optimization